Technical Session: Forecasting Applications for Large-Scale PV Operations
Moderator: Paul Stackhouse
PV Fleet Forecast Performance Evaluation
Hour to day-ahead PV fleet regional forecasts can be produced very accurately today, and provide ISOs with the means of producing accurate electrical load forecasts. Moreover, achieving 100% accuracy for regional forecasts can be accomplished at a lifetime operational cost amounting to much less than 10% of PV’s Capex.
Forecasting of Coastal Clouds in California
In summertime, marine boundary layer stratocumulus commonly cover the California coast and impact urban rooftop PV production. Accurate day-ahead solar forecasting of the associated ramp events promotes the economic operation of electricity markets under high PV penetration. Several advances in physical models related to coastal solar forecasting are presented.
Emerging opportunities in seasonal forecasting in support of renewable energy systems
Persistent weather patterns can cause renewable energy sources like incident sunlight and wind can vary over periods of weeks to months. Long-range forecasting can assist in managing energy supply and demand over these timescales.
Impact of Growing Distributed PV on Demand Load Forecasts
Sue Ellen Haupt
The National Center for Atmospheric Research has developed a forecast system that leverages weather information and historical data to provide short-range net electrical load forecasts. NCAR’s net load forecast uses a statistical learning approach based on regression trees that has proved to provide forecasts within a couple percent of accuracy on average. A benefit of this approach is the ability to integrate the solar power forecasts as an input to that load forecast. Such an integration of distributed PV with load forecasting will allow utilities and independent system operators to better deal with higher penetration of distributed generation.
Bias Correction of Satellite-Based Estimates of Irradiance Climatology for Hawaii Using Machine Learning Techniques
Improvements in Solar Power Forecasting: The Sunfast System